Jedpilled

Proverbs in Bio | Christ is Lord | Proverbs 9:8 | Ecclesia Semper Reformanda Est

[Jedpilled @jedpilled]

@hgafsd0 That's very funny Elias

[Elias @hgafsd0]

A sentence beginning and ending with boarding pass:
"Boarding, pass the boarding pass."

(the speaker is having dinner with Boarding and wants a boarding pass with his food)

[Jedpilled @jedpilled]

@Knight__dawg @Sarcastagasm_2 ftfy

[Breaking911 @Breaking911]

Footage allegedly shared by a U.S. soldier appears to show HIMARS missile launches targeting Iran from an undisclosed Gulf nation during the recent U.S.-Iran conflict.

[Jedpilled @jedpilled]

@W1GGLWRM @Pantera_pe @Paige_on_twt Read the study, they used ANOVA on the Trans dataset despite obvious non-normality and heteroscedasticity, it's very likely statistically non-significant or at least hasn't been analyzed properly.

[Adam Rochussen @AdamRochussen]

This is probably the worst p-hacking I’ve ever seen. Let’s go along with the validity of the “brain sex index”, for argument’s sake. The actual result shows that the men and the trans women did not significantly differ at all, but trans women and biological women did significantly differ. The authors just pretended otherwise via some insane statistical acrobatics.

They had three groups of 24 people each. The trans group is clearly not normally distributed (more on that later). They should have performed a Kruskal-Wallis test with Dunn’s multiple comparisons test. Instead, they did ANOVA and then *one-tailed* *unadjusted* t-tests for their post-hoc comparisons.

With this method, they found men vs trans women differed by d=0.64, p=0.016. But women vs trans women differed by d=1.87, p<0.001. So trans women are slightly more female than non-trans males. Narrative supported by data! Hooray! And to make it seem like it’s kind of half-way, they just give both of those p-values a single asterisk on the plot.

If they had instead done a two-tailed test, this doubles the p-value. And if they had done a Bonferroni correction for two comparisons, the p-value doubles again. They also should’ve done three comparisons really (men vs women isn’t tested in their paper—for no good reason), so triple instead of double the p-value.

So in reality, a proper analysis of their data without cherry-picking and biased hypothesis testing would’ve resulted in the difference between men and trans women being non-significant (p=0.096), but the difference between women and trans women remaining very significant (p<0.006). Narrative destroyed by data. Uh oh.

Going back to the squiggly distribution in the trans group. Of the 24 trans participants, 6 were androphiles (ie gay males), 18 were gynephiles (straight males). I don’t know for sure, because the authors didn’t share the underlying data, but it seems plausible that sexual attraction might be more important for their “brain sex index” than gender identity. We also don’t have sexual preference info for the non-trans groups, and in general the sample size is way too tiny to get fancy with multivariate analysis like this. It’s even too tiny for the main analysis: if someone wanted to replicate the d=0.64 difference properly, they’d need at least 53 people per group.

Anyways, if you want to support the wellbeing of trans people, which hopefully we all do, I don’t think twisting data to publish headline-grabbing fake conclusions is that way to go about it. It’s just going to make people irate when the statistical malpractice is pointed out, and then backfire on the trans “movement”. We really should stop attaching moral valence to data and stop trying to leverage the authority of science as a political tool.

[Jedpilled @jedpilled]

@unyxfly Self-centeredness is a good word, it resonates with its own generations too much but can "think" critically about other LLMs generations, makes combining multiple LLMs for problem solving more attractive, I know some work is being done there already.

[Jedpilled @jedpilled]

@unyxfly I've discovered this in experiments with LLMs solving math/logic problems, give a mistaken solution by Claude to Claude, it catches the mistake but makes another mistake, give a mistaken solution by Claude to GPT or vice versa, it finds the fix pretty easily.

[Jedpilled @jedpilled]

@reddit_lies It's actually an interesting view, because it effectively predicts society will divide into a number of internally-tolerant but externally-intolerant groups. Kind of like a leftist reinvention of the friend-enemy distinction.

[Jedpilled @jedpilled]

Sometimes people try to type "one" but they accidentally press "l" instead of "o" on their keyboard.

Thankfully, the meaning is unchanged.

[Jedpilled @jedpilled]

@upstatefederlst You could get it down to two weeks since the two radii will only overlap at two points, just check those two points and it'll be at one of them.

[Jedpilled @jedpilled]

@paleonormie The fact she keeps going back to the milk example proves exactly what it's about.

Because ??? Milk brands are basically the same? buy whatever? Unless you mean like skim milk or something in which case why would you buy skim milk? She's just a control freak.

[Jedpilled @jedpilled]

@reddit_lies All these political studies have a pretty big issue, which is that "Political Science" isn't a real science.

[Jedpilled @jedpilled]

@tenshiism Part of it might be that taller people are more noticeable in general. Plus tall people are more likely to be asked their height.

[Jedpilled @jedpilled]

@frank0daclonez That I cannot share.

[Earth @earthcurated]

Nearly 2,000 years ago today, Jesus Christ gave His life on the cross in Jerusalem.

[Jedpilled @jedpilled]

@Kruscle @VinciRSS Sounds great if you both live with the same timezone and schedules. eg.:

Person (2:49am): Hi?
Me (7:12am): Hello!
Person (10:34am): [The actual question or request]

They could've just written out the question at 2:49am so I could respond when I woke up and wasn't busy!

[Jedpilled @jedpilled]

@PstafarianPrice @reddit_lies I've heard stories (second hand, take with a grain of salt) that some are horrified and start accusing their doctors or family of stealing their balls, quite a bleak outcome, but these conditions are brutal regardless.

[Nick “The African Wild Dog” @killthedoctors]

your “based” right wing thought leader. I’ve been totally vindicated at this point. You all were worshipping this guy as the Hitler 2.0 and I was the only one who saw through it